cuSOLVERMp#
NVIDIA cuSOLVERMp is a high-performance, distributed-memory, GPU-accelerated library that provides tools for the solution of dense linear systems and eigenvalue problems.
cuSOLVERMp is compatible with 2D block-cyclic data layout and provides ScaLAPACK-like C APIs.
Download: cuSOLVERMp library is available through NVIDIA Developer Zone, NVIDIA HPC SDK, PyPI (CUDA 12, CUDA 13), and conda.
Key Features#
Multi-process, multi-GPU.
One process per GPU.
ScaLAPACK-like C functionalities and interfaces to facilitate porting.
NCCL communication backend.
Logging and tracing.
Tensor-core accelerated.
Index#
- Getting Started
- How to Use cuSOLVERMp
- Release Notes
- cuSOLVERMp v0.9.1
- cuSOLVERMp v0.9.0
- cuSOLVERMp v0.8.0
- cuSOLVERMp v0.7.2
- cuSOLVERMp v0.7.1
- cuSOLVERMp v0.7.0
- cuSOLVERMp v0.6.0
- cuSOLVERMp v0.5.1
- cuSOLVERMp v0.5.0
- cuSOLVERMp v0.4.3
- cuSOLVERMp v0.4.2
- cuSOLVERMp v0.4.1
- cuSOLVERMp v0.4.0
- cuSOLVERMp v0.3.1
- cuSOLVERMp v0.3.0
- cuSOLVERMp v0.2.0
- cuSOLVERMp v0.1.0
- Troubleshooting
- Software License Agreement